Min ROAS Works. Cost Per Result Doesn't. Here's What $1.5 Billion in Meta Spend Proves.

We spent years advising brands on cost-controlled bidding without being able to prove whether the platforms actually delivered on what they promised. That changed when we connected Claude to the Statlas database and ran the analysis at scale. What we found should change how you structure your Meta and Google accounts heading into Q4.

The Research: $1.5 Billion in Meta Spend Across 320 Accounts

Earlier this year, CTC's VP of Media Investment Tony Chopp ran the first major portfolio-wide analysis of Meta bidding strategy performance. The sample: 320 Meta accounts, just under $1.5 billion in spend, March through June 2026. The question was simple: do Meta's cost controls actually work?

The answer depends entirely on which cost control you use.

"Min ROAS is highly effective at achieving its target. Within several percentage points at the portfolio scale. Cost per result? Less than half the time." — Tony Chopp, VP of Media Investment

Min ROAS consistently hit its target across the portfolio. Cost per result goal, by contrast, achieved its target less than half the time. That is not a small gap. That gap shows up in your margin.

Why Min ROAS Outperforms Cost Per Result

The mechanics explain the performance difference. ROAS bidding gives the algorithm flexibility to go find high-value buyers, even if the cost per individual purchase is higher, because the return on the full order value still hits the target. Cost cap and cost per result bidding impose a strict ceiling on every conversion, which limits the system's ability to pursue higher-value buyers who happen to be more expensive to reach.

Think of it this way: ROAS bidding lets the algorithm optimize for order value. Cost cap bidding constrains the algorithm to a single acquisition cost threshold. The newer technology is more sophisticated, and the data shows it.

Google Confirmed the Same Pattern

CTC ran the same analysis on Google. The sample: 134 stores, just over $100 million in qualified spend, 24 months of data back to June 2024. The results aligned almost exactly with Meta.

Target ROAS bidding on Google ran above its target at a factor of 1.3x. Target CPA ran below its target at a factor of 0.8x. The pattern is consistent across both platforms: value-based bidding delivers. Cost-based bidding underdelivers.

"If you're going to build a media foundation, it should be ROAS-based bidding. Not because it sounds better. Because the data says it gives you the best chance to thread the needle." — Tony Chopp

The Auction Logic That Makes This Matter

The reason ROAS bidding strategy matters so much comes down to auction mechanics. The brand willing to spend the most to acquire a customer wins the auction. That means the strategic game is always the same: how low can you set your ROAS target and still make money?

The lower the target, the more aggressively you bid. The more aggressively you bid, the more auctions you win. The more auctions you win, the more customers you reach. But that math only holds if you can trust that the platform will actually deliver at the target you set.

For brands managing 7 to 9-figure media budgets, that confidence is the foundation of the entire paid media strategy. You are threading a needle between maximum investment and minimum viable return. One side of that needle produces contribution margin. The other torches it. The data now says ROAS-based bidding gives you the best chance of holding that line.

What Google Is Changing About Budget-Limited Campaigns

Google recently announced an update that is worth understanding before Q4. They are specifically addressing campaigns that show as "limited by budget" in the platform, and they are promising more consistent and predictable performance for those campaigns when budgets increase. If you have ever tripled the budget on a budget-limited Google campaign and watched the ROAS fall apart unexpectedly, this update is designed to solve that problem. It is a meaningful improvement to an otherwise frustrating experience with Google's bidding engine.

TikTok's Different Approach to the Same Problem

TikTok is approaching the budget liquidity problem from a different angle. In GMV Max campaigns within the TikTok Shop ecosystem, there is a budget scaling option that lets you set a baseline daily budget and define automated scale-up rules. If the system finds a pocket of demand and is hitting its target, it scales the budget up by a defined percentage, then resets the next day.

This solves the fundamental tension between what platforms want (budget liquidity, meaning large uncapped daily budgets) and what media buyers need (predictability, meaning confidence that the return will hold at scale). TikTok's approach lets you maintain budget control while still giving the algorithm room to find demand when it exists. We are starting to see this mechanism expand from GMV Max into standard web ads as well.

The CTC Canon: ROAS Bidding as Your Foundation

The data from this research has been formally incorporated into how CTC structures paid media across its portfolio. The conclusion is not that ROAS bidding works in every individual case — account-level variance is real and individual circumstances will always differ. The conclusion is that across the portfolio, ROAS-based bidding is consistently the right starting point. Cost caps and cost per result have a place in specific situations, but they should not be the foundation of your account structure.

Build your Meta and Google accounts on min ROAS and target ROAS bidding. Feed the algorithm quality conversion data through a strong Conversions API setup. Understand your incrementality so you know how low you can push the target before you cross into losing money. Then push it there. That is the CTC approach, and the data backs it up.

Frequently Asked Questions

Ready to Build Your Q4 Paid Media Foundation the Right Way?

CTC manages nearly $800 million in annual media investment across 170+ ecommerce brands. If you want a second opinion on your Q4 bid strategy before the season starts, talk to our team.

Talk to Us